Peroxisomal APX knockdown triggers antioxidant …Feb 06, 2014 · The physiological role of...

Original article

Peroxisomal APX knockdown triggers antioxidantmechanisms favourable for coping with highphotorespiratory H2O2 induced by CAT deficiency in rice

Rachel H. V. Sousa1, Fabricio E. L. Carvalho1, Carol W. Ribeiro2, Gisele Passaia2, Juliana R. Cunha1, Yugo Lima-Melo1,Márcia Margis-Pinheiro2 & Joaquim A. G. Silveira1

1Departamento de Bioquímica e Biologia Molecular, Universidade Federal do Ceará, Fortaleza, CE 60440-900, Brazil and2Departamento de Genética, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS 91501970, Brazil

ABSTRACT

The physiological role of peroxisomal ascorbate peroxidases(pAPX) is unknown; therefore, we utilized pAPX4 knock-down rice and catalase (CAT) inhibition to assess its role inCAT compensation under high photorespiration. pAPX4knockdown induced co-suppression in the expression ofpAPX3. The rice mutants exhibited metabolic changes suchas lower CAT and glycolate oxidase (GO) activities andreduced glyoxylate content; however, APX activity was notaltered. CAT inhibition triggered different changes in theexpression of CAT, APX and glutathione peroxidase (GPX)isoforms between non-transformed (NT) and silenced plants.These responses were associated with alterations in APX,GPX and GO activities, suggesting redox homeostasis differ-ences. The glutathione oxidation-reduction states weremodulated differently in mutants, and the ascorbate redoxstate was greatly affected in both genotypes. The pAPX suf-fered less oxidative stress and photosystem II (PSII) damageand displayed higher photosynthesis than the NT plants. Theimproved acclimation exhibited by the pAPX plants wasindicated by lower H2O2 accumulation, which was associatedwith lower GO activity and glyoxylate content. The suppres-sion of both pAPXs and/or its downstream metabolic andmolecular effects may trigger favourable antioxidant andcompensatory mechanisms to cope with CAT deficiency. Thisphysiological acclimation may involve signalling byperoxisomal H2O2, which minimized the photorespiration.

Key-words: Oryza sativa; ascorbate peroxidase; H2O2 homeo-stasis; oxidative stress; photorespiration; signalling.

INTRODUCTION

Despite the great advances in the understanding of oxidativemetabolism in plants in the last decades (to review see Foyer& Noctor 2011), several contradictory data involving the clas-sical antioxidant pathways have frequently been reported(Rizhsky et al. 2002; Miller et al. 2007; Koussevitzky et al.2008; Carvalho et al. 2014). In addition, the physiological role

displayed by some ascorbate peroxidase (APX) isoforms,especially peroxisomal isoforms, is poorly known. APX iscoded by a gene family that is widely represented in plants,and the existence of two peroxisomal APX (pAPX)isozymes, which are targeted for bind to the external surfaceof peroxisomal membranes, has been confirmed inArabidopsis (Panchuk et al. 2002; Narendra et al. 2006). Athird isoform possibly targeted to the peroxisomal matrix hasalso been reported (Panchuk et al. 2002). These isoformswere initially characterized many years ago (Yamaguchi et al.1995; Bunkelmann & Trelease 1996; Mullen et al. 1999). Inrice, two peroxisomal APX isoforms, OsAPX3 and OsAPX4,have been characterized, and both enzymes are also targetedfor membranes (Teixeira et al. 2004, 2006).

The most important remaining questions concern theimportance of pAPX in H2O2 scavenging duringphotorespiration and its possible role in signalling andperoxisome-cytosol cross-talking. The results reported todate remain controversial, but suggest that pAPX is impor-tant in the absence or in complement to catalase (CAT)(Wang et al. 1999). Works in which pAPX was overexpressedin Arabidopsis have suggested that transgenic plants aremore resistant to certain abiotic stresses (Kavitha et al. 2008),oxidative stress generated by CAT inhibition (Wang et al.1999) and methyl viologen (Kavitha et al. 2008). Tobaccotransgenic plants that overexpress pAPX from Salicorniabrachiata have displayed higher seed germination underosmotic and saline conditions (Singh et al. 2014). Neverthe-less, a study conducted with pAPX3 knockout (KO)Arabidopsis grown under normal growth conditions andexposed to salt, chilling and heat stresses clearly showed thatthis gene or the encoded protein is dispensable for plantgrowth and development (Narendra et al. 2006).

Peroxisomal APX isoenzymes may act in concert withother peroxidases in peroxisomes, especially CAT. Highphotorespiratory H2O2 production in C3 plants subjected toabiotic stress is common, and under extreme situations, thephotorespiration might represent up to 50% of Rubiscoactivity (Peterhansel & Maurino 2011). CAT has a high KM

for H2O2, whereas pAPX has high affinity for this substrate(Yamaguchi et al. 1995). The simultaneous activities of bothenzymes are important because they allow the elimination of

Correspondence: J. A. G. Silveira. Fax: 55 085-33669789; e-mail:[email protected]

Plant, Cell and Environment (2014) doi: 10.1111/pce.12409

bs_bs_banner

© 2014 John Wiley & Sons Ltd 1

mailto:[email protected]

high H2O2 amounts by CAT and the maintenance of lowH2O2 levels by APX activity. Indeed, APX enzymes are con-sidered to play a special role in the fine-tuning control ofH2O2 levels, allowing signalling for redox homeostasis andother physiological processes (Shigeoka et al. 2002). In con-trast, CAT activity has been associated with H2O2 scavengingand antioxidative protection in peroxisomes andglyoxysomes. In the presence of light and abiotic stresses,CAT degradation and/or delay in its re-synthesis may beinduced, resulting in a deficiency in its activity (Hertwig et al.1992).

Plants lacking CAT by KO, especially the Arabidopsis cat2mutant and others presenting CAT activity deficiency inducedby pharmacological inhibitors, are good models to elucidatethe role of pAPX and its involvement with CAT and otherantioxidants in peroxisomes.These plants display several phe-notypic anomalies, such as earlier senescence, followed bynecrotic spots on leaves, which are associated withphotorespiratory H2O2 accumulation (Chamnongpol &Willekens 1996; Willekens et al. 1997). The main oxidativefeature of these CAT-deficient plants is that the glutathione-oxidized form (GSSG) and total glutathione are stronglyincreased, and as a consequence, the redox state ofglutathione-reduced (GSH) is drastically decreased (Mhamdiet al. 2010). However, the biochemical mechanisms underly-ing the oxidation of the GSH pool under photorespiratoryH2O2 are not completely elucidated (Noctor et al. 2013). Inter-estingly, in Arabidopsis, tobacco and barley that are deficientin CAT, the ascorbate (ASC) pool remains practically unal-tered under high photorespiratory H2O2 (Kendall et al. 1983;Willekens et al. 1997), suggesting a minor antioxidant/signalling role.

Peroxisomal APX activity should play a central role inthe H2O2 control under CAT deficiency and in the presenceof normal activity of this enzyme. Unfortunately, few bio-chemical studies have characterized the effectiveness ofpAPX in peroxisomes, and the primary peroxidases thatoverlap with CAT activity are still unknown (Mhamdi et al.2010). However, plants have several other peroxidases thatcould reduce H2O2 to H2O, including the glutathioneperoxidase (GPX), phGPX, peroxidase-type glutathione-S-transferase (GST), thioredoxin-dependent peroxiredoxins,GST-dependent peroxiredoxins, glutaredoxin-dependentperoxiredoxins and type III peroxidases (Foyer & Noctor2009). Some of these peroxidases are localized in or out ofperoxisomes, and some require further confirmation (delRío et al. 2002; Schrader & Fahimi 2006). The relativeimportance of these enzymes for compensating and/or sup-plementing CAT and APX activities in peroxisomes isunclear (Kavitha et al. 2008).

The antioxidant metabolism in plants is complex, andexperimental evidence suggests the existence of redundantmechanisms, and unexpected results commonly occur.Indeed, tobacco and Arabidopsis double-KO mutants weremore resistant to oxidative stress than the single APX andCAT KO mutants (Rizhsky et al. 2002; Vanderauwera et al.2011). In addition, double-KO Arabidopsis mutants forAPX1 and chloroplast APX were also more resistant to oxi-

dative stress (Miller et al. 2007). Recently, our group, whilestudying rice with silenced cytosolic and chloroplastic APXs,also reported unexpected results. The deficiency of bothcytosolic and chloroplastic APXs greatly affected the redoxhomeostasis (Rosa et al. 2010), up-regulated a compensatorymechanism involving other peroxidases (Bonifacio et al.2011), triggered changes in the expression of genes and pro-teins not directly related to antioxidant metabolism (Ribeiroet al. 2012), and displayed a compensatory mechanism formaintaining photosynthesis (Carvalho et al. 2014) and amechanism for photoprotection complexes (Caverzan et al.2014).

Plant cells display strong coordination between chloro-plasts, peroxisomes, mitochondria and cytosol, especiallyduring photorespiration (Van Aken & Whelan 2012;Munné-Bosch et al. 2013). The complexity of these interac-tions connecting several cross-talk pathways may be able toexplain some of the complex and unexpected results con-cerning antioxidant enzymes localized in different cellularcompartments (Rizhsky et al. 2002; Miller et al. 2007;Vanderauwera et al. 2011). Changes in the H2O2 levels andoxidation-reduction states of ASC and GSH, especially GSHcontent, may act as powerful signals capable of generatingredox changes in compartments other than where they wereproduced (Mhamdi et al. 2010). For example, the excess H2O2

produced in cytosol caused by APX1 deficiency may triggeralterations in chloroplast metabolism (Davletova et al. 2005).In addition, the cross-talk may involve a transfer of reducingpower (NADH) between peroxisomes, chloroplasts, mito-chondria and cytosol, which would affect the redox state ofeach one of them (Yoshida et al. 2008).

Because the antioxidant physiological role of pAPX inplants is essentially unknown, especially in terms of its actionin photorespiratory H2O2 scavenging and its relationshipwith CAT deficiency, we conducted this study using an inte-grated molecular-physiological approach to assess theseunknown aspects. We utilized transgenic rice plants deficientin peroxisomal OsAPX4, which additionally presentednearly complete suppression of OsAPX3 expression. In thisstudy, we tested the following hypothesis: if pAPX is animportant peroxidase for H2O2 homeostasis in peroxisomes,then the rice mutants that are simultaneously deficient inboth pAPX and CAT enzymes should suffer higher oxidativestress induced by high photorespiratory H2O2 compared withnon-transformed (NT) plants.

The results obtained in this study do not support the pro-posed hypothesis. Instead, pAPX-silenced rice exhibited anantioxidant response that provided an improved acclima-tion to oxidative stress that was induced by highphotorespiration. The best performance in mutant plantswas confirmed by several physiological and biochemicalstress indicators (cell integrity, photosynthesis, lipidperoxidation and H2O2 accumulation). These responseswere associated with down-regulation of glycolate oxidase(GO) activity, higher CO2 assimilation and decreasedglyoxylate content. The possible physiological mechanismsinvolved with the acclimation of pAPX to CAT deficiencyare discussed later.

2 R. H. V. Sousa et al.

© 2014 John Wiley & Sons Ltd, Plant, Cell and Environment

MATERIALS AND METHODS

RNAi vector construct, plant transformation andthe growth of mutants and NT seedlings

A 218 bp sequence was amplified by PCR based on thesequence of the OsAPX4 gene (LOC_Os08g43560).The following primers were used: APx4RNAiF: 5′ AAAAAGCAGGCTCCTGACAAGGCATTGTTGGAAG 3′and APx4RNAiR: 5′ AGAAAGCTGGGTCCAGCTGCAGCAACAGCTACC 3′. The PCR product was cloned intothe pANDA vector (Miki et al. 2005), which enables hairpinstructure formation and post-transcriptional silencing of theOsAPX4 gene. The pANDA vector contains the maizeubiquitin promoter and the Hpt gene for selectionby hygromycin. The transformation of rice calli was achievedvia Agrobacterium tumefaciens as described previously(Upadhyaya et al. 2000). Regenerated seedlings were grownat 28 °C in 10× diluted Murashige and Skoog (MS) mediumwith a photoperiod of 12 h and 150 μmol m−2 s−1 of photosyn-thetic photon flux density (PPFD) in a growth chamber for7 d. Three lines of pAPX4 transgenic plants were selectedbased on the transcript amounts: Lg, Lh and Lj, whichexpressed 10, 15 and 14%, respectively, of the transcriptamount in the NT plants.

Plant growth in the greenhouse

The Lg, Lh and Lj lines and the NT plants were transferred to3 L plastic pots filled with half-strength Hoagland–Arnon’snutritive solution (Hoagland & Arnon 1950). The pH wasadjusted to 6.0 ± 0.5 every 2 d, and the nutrient solution waschanged weekly. The seedlings were previously grown for45 d in a greenhouse under natural conditions: day/nightmean temperature of 29/24 °C and mean relative humidity of68% and a photoperiod of 12 h. The light intensity (fromsunlight) inside the greenhouse varied as it would for atypical day from 0600 to 1800 h, reaching an average ofmaximum PPFD of 820 μmol m−2 s−1 at noon.

3-Amino-1,2,4-triazole (AT) applicationto entire plants

Forty-five-day-old transformed and NT plants weregrown as previously described. For the entire-plantexperiment, a group of plants was transferred to a growthchamber at 27 ± 2 °C/24 ± 2 °C (day/night), 70 ± 10% humid-ity, 400 μmol m−2 s−1 PPFD and a 12 h photoperiod. Theplants were acclimated under these conditions for 24 h.After this time, the specific CAT inhibitor (10 mM AT dis-solved in 10 mM N-(2-Hydroxyethyl)piperazine-N′-(2-ethanesulfonic acid) (HEPES) buffer at pH 6.5 with 1.5 mMCaCl2 and 0.1% Triton X-100) was sprayed in excess on theshoot until a complete wetting was achieved.The plants weretreated for 12 h, and after AT application, they did not exhibitany visual symptoms of leaf toxicity. The control plants weresprayed with the same solution without AT. The leaf sampleswere immediately harvested, separated for physiological, bio-chemical and PCR analysis, placed in liquid N2 and trans-ferred to a −85 °C freezer.

Leaf segment experiments: CAT inhibition, GOinhibition and high light application

For the assay with leaf segments, the plants were previouslyacclimated in a growth chamber at the same conditionsand for the same time period described for the intactplants experiments. For the AT treatment, leaf segments(10 cm) from 45-day-old plants were immersed in a bufferof 10 mM HEPES at pH 6.5 that contained CaCl2 1.5 mMand 0.01% Triton X-100 (v/v) in the presence of 5 mM AT.For the GO inhibition treatment, 10 mM α-hydroxy-2-pyridinemethanesulfonic (HPMS) was dissolved in the samemedium utilized for AT. The leaf segments remainedimmersed in the 10 mM HPMS for 24 h or 12 h (time-course). In the low- and high-light experiments, the leaf seg-ments were immersed in the same solution utilized for theCAT and GO inhibition using the previously described pro-cedure.The plates were divided into four groups: (1) low light– LL (200 μmol m−2 s−1); (2) LL + 5 mM AT; and (3) high light– HL (1000 μmol m−2 s−1) + 5 mM AT. In treatments 1 and 2,the AT and LL were supplied for 24 h, whereas in treatment3, the AT was supplied for 12 h followed by 12 h of HLapplication. Treatment 1 was used as the control. The leafsegments were collected using the same procedure as for theentire plant.

Determining gas exchange and chlorophylla fluorescence

The net CO2 assimilation rate (PN), transpiration andstomatal conductance (gs) were measured using a portableinfrared gas analyser (IRGA) system equipped with a light-emitting diode (LED) source and leaf chamber (IRGALI-6400XT, LI-COR®, Lincoln, NE, USA). The IRGAchamber had the following internal parameters:1000 μmol m−2 s−1 PPFD, 1.0 ± 0.2 kPa vapour pressure deficit(VPD), 38 Pa CO2 and 28 °C. In vivo chlorophyll a fluores-cence was measured using a LI-6400-40 leaf chamberfluorometer (LI-COR) coupled with the IRGA equipment.The actinic light utilized for the measurement of gasexchange and chlorophyll a fluorescence was 1000 PPFD,which corresponded to the saturation light for CO2 assimila-tion in rice (Carvalho et al. 2014). The fluorescence param-eters were measured using the saturation pulse method(Schreiber et al. 1994) in leaves exposed to light and 30 mindark-adapted conditions. The intensity and duration of thesaturation light pulse were 8000 μmol m−2 s−1 and 0.7 s,respectively. The amount of blue light was set to 10% of thePPFD to maximize the stomatal aperture (Flexas et al. 2007).The following parameters were assessed: the maximumquantum yield of photosystem II (PSII) [Fv/Fm = (Fm − Fo)/Fm], which was measured under 30 min dark-adapted condi-tions, and the effective quantum yield of PSII [ΔF/Fm′ = (Fm′ − Fs)/Fm′], which was measured in leavesexposed to actinic light (1000 μmol m−2 s−1 PPFD). The pho-tochemical quenching coefficient was calculated as follows:qP = (Fm′ − Fs)/(Fm′ − Fo). The Fm and Fo parameters cor-respond to the maximum and minimum fluorescence of the

pAPX KD rice displays high H2O2 acclimation 3


dark-adapted leaves, respectively. Fm′ and Fs are themaximum and steady-state fluorescence in the light-adaptedleaves, respectively, and Fo′ is the minimum fluorescenceafter the far-red illumination of the previously light-exposedleaves (Schreiber et al. 1994; Flexas et al. 2007).

Electrolyte leakage and lipid peroxidation

Membrane damage (MD) or cellular viability was measuredby electrolyte leakage as described previously by Blum &Ebercon (1981). Twenty leaf discs (1.0 cm diameter) wereplaced in test tubes containing 20 mL of deionized water.Theflasks were incubated in a shaking water bath (25 °C) for12 h, and the electrical conductivity in the medium (L1) wasmeasured. The discs were then boiled (95 °C) for 60 min andcooled to 25 °C, and the electrical conductivity (L2) wasmeasured again. The relative MD was estimated byMD = L1/L2 × 100. The lipid peroxidation was measuredbased on the formation of thiobarbituric acid-reactive sub-stances (TBARS) in accordance with Cakmak & Horst(1991). The concentration of TBARS was calculated using itsabsorption coefficient (155 mM−1 cm−1), and the results areexpressed as ηmol MDA-TBA g FW−1.

Determination of the ASC-GSH redox state

The contents of reduced (ASC), oxidized (DHA) and totalascorbate (ASC + DHA) were determined according to theprotocol described by Queval & Noctor (2007). The assay ofASC was based on decrease of the ASC absorbance at265 nm in the presence of ASC oxidase (AO). For determi-nation of total ASC (ASC + DHA), the DHA was reducedby 25 mM DTT, and total ASC was measured as described forreduced ASC. The DHA content was calculated by the dif-ference between total ascorbate and ASC, and all forms wereexpressed as μmol g fresh matter (FM)−1 calculated from anASC standard curve. GSH was measured by the GSH reduc-tase (GR)-dependent reduction of 5,5′-dithio-bis (2-nitro-benzoic acid), DTNB, according to the Griffith (1980)method and detailed protocol described in Queval & Noctor(2007). The GSH and GSSG content were expressed asμmol g FM−1 and calculated from standard curves of GSHand GSSG, respectively.

Diaminobenzidine (DAB) staining

In situ detection of peroxide (H2O2) was performedby staining with 3,3′-DAB as previously described(Thordal-Christensen et al. 1997). Leaf segments were infil-trated under dark vacuum conditions with 10 mM potassiumphosphate buffer, 10 mM NaNO3 and 0.1% (w/v) 3,3′-DAB,pH 7.8. The segments were incubated for approximately 16 hin dark conditions and then destained with 15% (w/v)trichloroacetic acid (4:1 v/v) ethanol : chloroform for 48 hbefore being photographed.

qRT-PCR

qRT-PCR experiments were conducted using cDNA synthe-sized from total RNA purified with Trizol (Invitrogen®,

Carlsbad, CA, USA) as previously described (Rosa et al.2010). The primer pairs to amplify the Osfdh3 gene(LOC_Os02g57040) and Osfa1 gene (LOC_Os03g08020)were used as internal controls to normalize the amount ofmRNA present in each sample. All qRT-PCR reactions wereperformed with a StepOne plus real-time PCR system(Applied Biosystems®, Foster City, CA, USA) using SYBR-green intercalating dye fluorescence detection.

Enzymatic activity assays

All enzymatic activities were assayed from the total solubleprotein extract. The protein extraction was performed utiliz-ing fresh leaf matter in the presence of 100 mM phosphatebuffer, pH 7.5, containing 2 mM EDTA and 1 mM ASC. TheAPX activity was assayed following ASC oxidation by meas-uring the decrease in absorbance at 290 nm (Amako et al.1994). The APX activity was expressed as μmol ASCmg protein−1 min−1.The CAT activity was measured by follow-ing the oxidation of H2O2 at 240 nm over a 300 s interval at25 °C in the presence of 50 mM potassium phosphate buffer,pH 7.0, containing 20 mM H2O2 (Havir & McHale 1987).TheCAT activity was calculated according to the molar extinctioncoefficient of H2O2 (36 × 10−3 mM−1 cm−1) and expressed asμmol H2O2 mg protein−1 min−1. The superoxide dismutase(SOD) activity was determined by inhibiting blue formazanproduction via Nitro Blue Tetrazolium (NBT)photoreduction. The reaction was started by illumination(30 W fluorescent lamp) at 25 °C for 6 min, and the absorb-ance was measured at 540 nm (Giannopolitis & Ries 1977).One SOD activity unit (UA) was defined as the amount ofenzyme required to inhibit 50% of NBT photoreduction, andthe activity was expressed as UA mg protein−1 min−1

(Beauchamp & Fridovich 1971). The GO activity andglyoxylate concentration were assayed by the formation of theglyoxylate–phenylhydrazone complex and measured afterreading at 324 nm (Baker & Tolbert 1966). GO activity wascalculated using the molar extinction coefficient of theglyoxylate-phenylhydrazone complex (17 mM−1 cm−1) andexpressed as ηmol glyoxylate−1 mg protein−1 min−1, whereasthe glyoxylate concentration was determined from a standardcurve. The GPX activity was assayed by measuring the GSHoxidation using cumene hydroxide as an electron acceptor.The method is based on the measurement of GSH oxidationby the Griffith (1980) method after correction of GSH oxida-tion in the absence of cumene hydroxide. The enzymaticreaction medium was composed of 100 mM phosphate bufferat pH 7.0, which contained cumene hydroxide 0.5 and 4.0 mMGSH in the final reading medium. The reaction was stoppedafter 3 min at 25 °C.The GPX activity was expressed in ηmolGSH mg protein−1 min−1. The GST activity was assayed bydirect GSH consumption as described by Valentovicová et al.(2009), and the activity was expressed in ηmol GSHmg protein−1 min−1.

Statistical analysis

The experiments were conducted in a 2 × 2 or 2 × 2 × 2 fac-torial design with two genotypes and two AT concentrations



or two genotypes, two AT concentrations and two lightregimes. An individual pot containing two plants or an indi-vidual Petri plate containing 15 leaf segments representseach replicate. Each experiment had at least four replicates.All the measurements were performed with at least fourreplicates from independent pots and plates. The data wereanalysed by analysis of variance (anova), and the meanswere compared using Tukey’s test (P < 0.05).

RESULTS

Physiological, biochemical and molecularcharacterization of the APX4knockdown mutants

In this study, three rice lines (Lg, Lh and Lj) silenced in theperoxisomal APX4 were initially characterized to identifypossible phenotypic changes after APX4 silencing. None ofthe studied lines displayed an altered phenotype from thegermination to vegetative development phase (45 d after ger-mination), which is shown in Supporting Information Fig. S1.However, pAPX4 silencing nearly completely suppressedOsAPX4 gene expression in all the selected rice lines(Table 1), indicating that hairpin expression was very effec-tive. As an indicator of plant growth, the fresh shoot matteraccumulated at 45 d was similar between the pAPX knock-down lines and NT plants; this result is in contrast to the netphotosynthesis (PN), which was slightly higher in the APX4lines compared with the NT plants.The stomatal conductancedisplayed a trend similar to PN in all the studied plants. Theanalysed parameters associated with PSII efficiency exhib-ited a different trend compared with the gas exchangeparameters. The actual quantum yield of PSII (ΔF/Fm′) andelectron transport rate (ETR) at the PSII level did notchange in any of the lines compared with the NT plants(Table 1). The APX, CAT, SOD and GO activities were

evaluated to investigate whether the APX4-silenced plantsexhibited changes in antioxidant enzymes and a key proteininvolved with photorespiration. The APX activities in thesilenced lines were similar, CAT was lower, SOD was similar,and GO was significantly decreased compared with the NTplants (Table 2). After the biochemical and physiologicalcharacterization of these mutant lines, Lg was selected torepresent the APX4 mutants because it displayed highersilencing and similar physiological and biochemical param-eters compared with the other lines.

CAT inhibition differentially affects theexpression of the OsAPX, OsGPX and OsCATisoforms in NT and APX4 mutants

The APX-silenced plants triggered an almost complete sup-pression of OsAPX3 gene expression (a reduction ofapproximately 90% in the transcript amount compared withNT), which was most likely a result of the high homologydisplayed by these two genes (Fig. 1). Thus, the APX4mutants used here were deficient in both peroxisomal APXs.In contrast, the quantity of transcripts of other OsAPX andOsGPX isoforms was slightly changed in the APX-silencedmutants in the control condition (Fig. 2a,b). To assess thecombined effects of pAPX and CAT deficiency, an experi-ment was performed that utilized the classical 10 mM 3-ATCAT inhibitor, which was sprayed on the rice shoots of45-day-old plants for a short period of time (12 h). The ATtreatment did not change the transcript amount of eitherAPX4 or APX3 in the mutants, which remained reduced by90% compared with the NT control. This treatment also didnot alter the expression of these genes in the NT plants. Inthe presence of AT, the cytosolic OsAPX2 was up-regulatedin both genotypes, but was more strongly expressed in the NTplants, which displayed a sixfold increase compared with thecontrol. In contrast, higher transcript amounts of the

Table 1. Transcript amounts of OsAPX4 and physiologicalcharacterization of three pAPX4-silenced rice lines and NT plantsafter 45 d of growth in nutrient solution

Parameters

Lines

NT Lg Lh Lj

OsApx4 transcriptamounts

1.000a 0.090b 0.146b 0.141b

Shoot FW (g per plant) 30.83a 28.76a 27.12a 26.71aPN ( μmol m−2 s−1) 19.40b 25.56a 22.51a 25.39ags (mol m−2 s−1) 0.17b 0.31a 0.25a 0.33aϕPSII (ΔF/Fm′) 0.29a 0.32a 0.32a 0.28aETR ( μmol m−2 s−1) 126.58a 144.38a 142.3a 122.89a

Letters represent significant differences between the mutant linesand NT plants within each parameter at a confidence level of 0.05.The data are presented as the means of four replicates (±sd) andwere compared by Tukey’s test.ETR, electron transport rate; FW, fresh weight; NT, non-transformed; pAPX, peroxisomal ascorbate peroxidises; PSII,photosystem II.

Table 2. Activities of enzymes in the leaves of threepAPX4-silenced rice lines and NT plants after 45 d of growth innutrient solution

Antioxidant enzymes

Lines

NT Lg Lh Lj

APX (μmol ASCmg prot−1 min−1)

0.56a 0.52a 0.54a 0.53a

CAT (μmol H2O2

mg prot−1 min−1)74.02a 66.58ab 59.28b 55.52b

SOD (U mg prot−1 min−1) 0.91a 1.04a 0.96a 0.97aGO (ηmol glyoxylate

mg prot−1 min−1)91.22a 71.48b 70.17b 70.06b

Letters represent significant differences between the mutant linesand NT plants within each parameter at a confidence level of 0.05.The data are presented as the means of four replicates (±sd) andwere compared by Tukey’s test.APX, ascorbate peroxidise; ASC, ascorbate; CAT, catalase; SOD,superoxide dismutase; GO, glycolate oxidase; pAPX, peroxisomalascorbate peroxidises; PSII, photosystem II.



chloroplastic OsAPX7 were observed in the pAPX4 mutants,increasing by nearly fourfold compared with the NT control.The OsAPX5 and OsAPX6 transcript amounts (bothmitochondrial) displayed a slight increase in the AT-treatedAPX4 plants compared with NT control (Fig. 2a). OsAPX4silencing did not alter the expression of the five OsGPXgenes in the control condition. The AT treatment triggered aslight increase in the amount of OsGPX1 and OsGPX2 tran-scripts (mitochondrial and cytosolic GPXs, respectively),whereas OsGPX3 (mitochondrial) increased by threefold inthe transgenic plants and increased slightly in the NT com-pared with NT control. CAT inhibition did not alter OsGPX4expression (chloroplastic) in either the NT or silenced plants,but induced a slight increase in OsGPX5 (cytosolic) expres-sion compared with the NT plants (Fig. 2b). Rice has threeCAT genes: OsCATA, OsCATB and OsCATC. The silencingof OsAPX4 did not change the OsCATA transcript amountin the mutants, but it was slightly down-regulated in the pres-

ence of AT (Supporting Information Fig. S2). The AT treat-ment down-regulated OsCATA more intensely in NT plants.Compared with OsCATA, the OsCATB gene was slightlyup-regulated in the silenced mutants; however, CAT inhibi-tion triggered a strong up-regulation of OsCATB in themutants. The OsCATB gene was up-regulated by CAT inhi-bition in the NT plants to a lower extent compared with themutants (Supporting Information Fig. S2). The OsCATCgene was shown to encode a protein localized in the photo-synthetic tissue and related to photorespiration (Iwamotoet al. 2000). In this study, the AT completely inhibited thetotal CAT activity in the leaves of both genotypes, which willbe shown later.

Changes in cell integrity and photosynthesisthat indicate that the APX4 mutants exhibitedbetter acclimation to CAT inhibition at theentire-plant level

In this experiment, the entire rice plants (45-day-old) weresprayed with 10 mM AT for 12 h (a short-term exposure) to

Figure 1. Transcript levels of the OsAPX3 and OsAPX4 genes inNT and APX4 leaves of rice plants exposed to control or 10 mMAT. Different lowercase letters represent significant differencesbetween the treatments within lines (NT and APX4), and differentcapital letters represent significant differences between the lineswithin treatments (control, AT) at a confidence level of 0.05. Thedata are presented as the means of four replicates (±sd) and werecompared by Tukey’s test. AT, 3-amino-1,2,4-triazole; NT,non-transformed.

Figure 2. Transcript levels of the OsApx (a) and OsGpx(b) genes in NT and APX4 leaves of rice plants exposed to controlor 10 mM AT. Different lowercase letters represent significantdifferences between the treatments within lines (NT and APX4),and different capital letters represent significant differencesbetween the lines within treatments (control, AT) at a confidencelevel of 0.05. The data are presented as the means of fourreplicates (±SD) and were compared by Tukey’s test. AT,3-amino-1,2,4-triazole; NT, non-transformed.



induce a complete inhibition of CAT activity in leaves whileminimizing other physiological side effects caused by excessAT. After the 12 h treatment, both photosynthesis (CO2

assimilation) and cell integrity decreased more intensely inthe NT plants compared with the pAPX mutants (Figs 3a and4a). These effects were proportional to the AT exposure timefor both genotypes (data not shown).The changes induced bythe transient CAT deficiency suggested that these symptomsreflected a mild physiological stress in both the NT andsilenced mutants because the PN and electrolyte leakage inthe leaves changed slightly. The potential quantum yield ofPSII and Fv/Fm decreased slightly in both plants exposed toAT, indicating that the PSII apparatus was essentially unaf-fected by CAT inhibition (Fig. 3b). The TBARS content, anindicator of lipid peroxidation, was slightly increased in bothgenotypes; these minor changes apparently did not reflect anoxidative stress, and they corroborated the results that wereobtained with other stress indicators and indicated that CAT

inhibition for a short-time period induced slight physiologicaldisturbances (Fig. 4b).

CAT inhibition produced different effects on theactivities of antioxidant enzymes and GO in theattached leaves of silenced and NT plants

To analyse the effects of CAT inhibition on the activitiesof other antioxidant enzymes and an important photo-respiration enzyme, the activities of CAT,APX, GPX, GST andGO were measured in the leaves of both genotypes. Despite ashort-term exposure to theAT inhibitor (12 h),CAT activity wascompletely inhibited after 12 h in both of the studied genotypes(Fig. 5a). This AT-induced CAT deficiency was sufficient toinduce significant changes in the oxidative and antioxidant enzy-matic activities. APX activity was not altered in the NT plants,whereas in the mutants, theactivity was increased in response to

Figure 3. (a) Net photosynthesis and (b) maximum quantumyield in NT and APX4 leaves of rice plants exposed to control or10 mM AT. Different lowercase letters represent significantdifferences between the treatments within lines (NT and APX4),and different capital letters represent significant differencesbetween the lines within treatments (control, AT) at a confidencelevel of 0.05. The data are presented as the means of fourreplicates (±sd) and were compared by Tukey’s test. AT,3-aminotriazole; NT, non-transformed.

Figure 4. Changes in (a) electrolyte leakage percentage and(b) lipid peroxidation in NT and APX4 leaves of rice plantsexposed to control or 10 mM AT. Different lowercase lettersrepresent significant differences between the treatments withinlines (NT and APX4), and different capital letters representsignificant differences between the lines within treatments (control,AT) at a confidence level of 0.05. The data are presented as themeans of four replicates (±sd) and were compared by Tukey’s test.AT, 3-amino-1,2,4-triazole; NT, non-transformed.



CAT inhibition (Fig. 5b).GPX activity was increased in responseto CAT deficiency in both plants, but this increase was higher inthe mutants compared with NT plants (Fig. 5c).GST activity washigher in NT than in mutants under both experimental condi-tions,butAT induced a slight decrease in GST activity only in theNT plants (Fig. 5d).GO activity remained essentially unchangedin both genotypes treated with AT, but the peroxisomal APXmutants exhibited lower activity under both experimental con-ditions (Fig. 6a), indicating a potential to display lowerphotorespiratoryH2O2 productioncomparedwiththeNTplants.In the presence of HPMS, a specific GO inhibitor, a close corre-lation between GO activity (Supporting Information Fig.S3) andCAT activity (Fig. 6b) was observed. In addition, in the APX4mutants, CAT activity was lower than in the NT plants underboth experimental conditions. Thus, GO inhibition induced a

parallel decrease in CAT activity;however,when CAT was inhib-ited, the GO activity did not change (Fig. 6a).These results mayindicate that GO activity is essential for the control ofperoxisomal H2O2 levels via glycolate oxidation, which is associ-ated with the level of CAT activity; however, reciprocal regula-tion between these two enzymes could not occur in peroxisomes.This close relationship between these two enzymes was alsonoted under control conditions jointly with glyoxylate content,the product of the GO reaction (Fig. 6c). Unexpectedly, theglyoxylate content was only strongly decreased in NT plantsbecause of CAT inhibition. The glyoxylate content wasunchanged in the mutants in response to AT, but it exhibitedlevels similar to those noted in the NT plants exposed to CATinhibition. Interestingly, a close relationship between GO,glyoxylate and CAT activity was observed in the two genotypes

0

0

25

50

75

30

60

CAT

activity

(μm

ol H

2O

2 m

g p

rote

in−1

min

−1)

GP

X a

ctivity

(ηm

ol G

SH

mg p

rote

in−1

min

−1)

AP

X a

ctiv

ity

(μm

ol A

SC

mg p

rote

in−1 m

in−1)

GS

T a

ctiv

ity

(ηm

ol G

SH

mg p

rote

in−1 m

in−1)

NT

aA

aA

aA

aA

aA

aB

aB

aB aB

aB

bB

bA

bA

bA

bA

bA

APX4 NT APX4

90

120

0.0

0

3

6

9

0.2

0.4

0.6

0.8

Control

AT

(a) (b)

(c) (d)

Figure 5. (a) CAT, (b) APX, (c) GPX and (d) GST activities in the NT and APX4 leaves of rice plants exposed to control or 10 mM AT.Different lowercase letters represent significant differences between the treatments within lines (NT and APX4), and different capital lettersrepresent significant differences between the lines within treatments (control, AT) at a confidence level of 0.05. The data are presented as themeans of four replicates (±sd) and were compared by Tukey’s test. AT, 3-amino-1,2,4-triazole; APX, ascorbate peroxidise; CAT, catalase;GPX, glutathione peroxidase; GST, glutathione-S-transferase; NT, non-transformed.



undercontrolconditions,butnot in thepresenceofAT,indicatingthat under CAT inhibition, the relationship between the activ-ities of these two enzymes are differently modulated in the twostudied genotypes.

Peroxisomal APX-silenced mutants displayedchanges in the oxidation-reduction states ofascorbate and glutathione under CAT inhibitionat the entire-plant level

The oxidation-reduction states of ascorbate and glutathionein the presence and absence of CAT inhibition in APX4mutants was evaluated to investigate whether the silencingcould trigger changes in these non-enzymatic antioxidantsand their redox states. The total ascorbate content and ASCredox state were slightly lower (70%) in the mutants com-pared with NT plants (80%), in the control condition(Fig. 7a). CAT inhibition induced a strong decrease in theASC redox state to approximately 26 and 30% in the NT andmutant plants, respectively, but the total ascorbate was notaltered in either genotype. These data may indicate that ASCwas intensely oxidized to DHA under CAT inhibition and

Figure 6. (a) GO, (b) CAT activities and (c) glyoxylate contentin the NT and APX4 leaves of rice plants exposed to control or10 mM AT or 10 mM HPMS. Different lowercase letters representsignificant differences between the treatments within lines (NT,HPMS and APX4), and different capital letters representsignificant differences between the lines within treatments (control,AT) at a confidence level of 0.05. The data are presented as themeans of four replicates (±sd) and were compared by Tukey’s test.AT, 3-amino-1,2,4-triazole; APX, ascorbate peroxidise; CAT,catalase; GO, glycolate oxidase; HPMS,α-Hydroxypyridinemethanesulfonic acid; NT, non-transformed.

Figure 7. (a) ASC and (b) GSH contents in the NT and APX4leaves of rice plants exposed to control or 10 mM AT. Differentlowercase letters represent significant differences between thetreatments within lines (NT and APX4), and different capitalletters represent significant differences between the lines withintreatments (control, AT) at a confidence level of 0.05. The data arepresented as the means of four replicates (±sd) and werecompared by Tukey’s test. AT, 3-amino-1,2,4-triazole; APX,ascorbate peroxidise; ASC, ascorbate; GSH, glutathione; NT,non-transformed.



probably high peroxisomal H2O2 or that secondary oxidizingconditions altered it into peroxisomes.These changes in ASCmetabolism occurred to a similar extent in both plant types.The NT and mutant plants presented similar values for totalglutathione, GSH and GSSG in the control condition.However, AT treatment differentially and remarkablyaltered the GSH oxidation-reduction states in these geno-types. In the NT plants, total glutathione was increased byfivefold, whereas in the mutants, total glutathione wasincreased by 2.5-fold compared with the control (Fig. 7b).The GSH content was not changed in the NT plants, butdecreased by 50% in the mutants. In contrast, the GSSGcontent was increased similarly in both genotypes. The NTplants exhibited higher GSH redox state (42%) comparedwith mutants (28%). These results indicate that CAT inhibi-tion induced higher GSH synthesis in the NT plants andallowed a more favourable GSH redox state compared withthe APX4 mutants. However, these changes in the GSHredox state strongly suggest that CAT inhibition inducedhigh photorespiratory H2O2 production; this reaction hasbeen widely observed in Arabidopsis cat2 mutants, which arethe major model for studies on the deficiency of CAT andperoxisomal H2O2 accumulation.

APX4-silenced rice displayed better acclimationto oxidative stress induced under highphotorespiration as modelled by leaf segments

In the experiments with leaf segments, both the NT plants andmutants were exposed to CAT inhibition combined withenhanced photorespiration induced by the high-light condi-tion to compare the differential acclimation between the twogenotypes in response to an acute oxidative stress generatedby high peroxisomal H2O2 levels. CAT inhibition, in the pres-ence of high light, initially caused yellow spots in the leafsegments, followed by generalized senescence. These effectswere remarkably more accentuated in the NT plants than theAPX mutants (Fig. 8a). Interestingly, the single AT treatmentunder low light or single high-light conditions (data not

shown) did not cause any injury symptoms, indicating thatsenescence was induced by CAT deficiency combined with theenhanced photorespiration induced by high-light conditions.These strong injury symptoms were closely correlated withintense H2O2 accumulation in the leaf tissue, as revealed byDAB staining (Fig. 8b). The single AT treatment in presenceof low light also caused slight H2O2 accumulation, which washigher in the NT plants than the mutants and H2O2 accumu-lated preferentially in the leaf midrib (Fig. 8a,b). The visualindications and H2O2 accumulation intensity were correlatedwith important indicators of physiological and oxidativestresses, which are shown in Fig. 9.The electrolyte leakage (anindicator of cellular integrity) and TBARS content (an indi-cator of lipid peroxidation) in the NT plants were 90 and 100%higher than in the mutants, respectively, whereas the Fv/Fm(an indicator of PSII integrity), was 130% higher in themutants than in the NT plants when both were exposed to thecombination of AT and high light (Fig. 9).These data stronglyindicate that the APX4 mutant was remarkably more accli-mated than the NT plants to an acute oxidative stress inducedby high photorespiratory H2O2 caused by CAT deficiencycombined with high photorespiration.

DISCUSSION

The data obtained in this study are intriguing; rice mutantsdeficient in both pAPX isoforms were less susceptible to mildoxidative stress induced by CAT inhibition under a moderatelight regime (400 μmol m−2 s−1) and high-light regime(1000 μmol m−2 s−1). Our results are surprising because severalstudies utilizing transgenic plants that overexpress pAPX havedemonstrated that these mutants are more resistant to differentabiotic stresses (Wang et al. 1999;Li et al. 2009;Singh et al. 2014).Contrary to these reports, Narendra et al. (2006) utilizedperoxisomal-APX3 KO in Arabidopsis and demonstrated thatthis protein is dispensable for development in normal growthconditions and under abiotic stress. Intriguingly, there are nopublished studies concerning the performance of pAPX-deficient mutants (using knockdown, KO or inhibitors) inresponse to high photorespiratory H2O2 conditions, which wasinvestigated here. Thus, our report is the first to clearly demon-strate that pAPX deficiency in rice plants is likely capable oftriggering an effective biochemical/molecular compensatorymechanism for coping with high peroxisomal H2O2. Paradoxi-cally, the performance exhibited by these plants was inconsistentwith expectations because the deficiency in both pAPX and CATin the mutants should have induced a higher sensitivity to oxida-tive stress compared with that of the NT plants.

The complex and unexpected results displayed by pAPXmutants in response to oxidative disturbances are similar toother results involving simultaneous deficiency of cytosolicAPX1 and CAT in tobacco and Arabidopsis (Rizhsky et al.2002; Vanderauwera et al. 2011), cytosolic and thylakoid APXin Arabidopsis (Miller et al. 2007), cytosolic APX1/APX2 inrice (Rosa et al. 2010) and cytosolic APX1/APX2 inArabidopsis (Suzuki et al. 2013). Unfortunately, despite ofthe existence of these data involving simultaneous down-expression of APX and/or CAT in diverse species and

Figure 8. Appearance of leaf segments (a) and in situlocalization of hydrogen peroxide (b) using DAB dye in NT andAPX4 leaf segments of rice plants exposed to control or 10 mMAT with or without the increased light regime. The pictures wereselected as the most representative from four independentreplicates. AT, 3-amino-1,2,4-triazole; APX, ascorbate peroxidise;DAB, diaminobenzidine; NT, non-transformed.



distinct cellular compartments, these complex responses arestill debated, and the elucidation of biochemical mechanismsis lacking. Mittler et al. (2011) and Carvalho et al. (2014) pre-sented an interesting hypothesis suggesting that for each spe-cific oxidative condition, a singular antioxidant responsecould occur. In addition, these authors have suggested thatthe classical antioxidant pathways should be revised and thatnew metabolic models are needed to explain these complexresponses, especially in other model plants in addition toArabidopsis, such as rice.

An overall explanation of the results obtained in this studyis that pAPX mutant rice plants could have displayed largeepigenetic changes and high phenotypic plasticity. Thesechanges could have been related to efficient compensatoryantioxidant mechanisms designed to cope with specific oxida-tive challenges, which has already been reported for riceplants silenced for cytosolic APX1/2 (Rosa et al. 2010;Bonifacio et al. 2011; Ribeiro et al. 2012; Carvalho et al. 2014)and chloroplasticAPX (Caverzan et al. 2014). In principal, thepAPX silencing should have caused a significant decrease inthe peroxisomal APX activity, which should have result inchanges in the peroxisomal H2O2 metabolism. These changescould have triggered additional antioxidant pathways, pos-sibly via H2O2 and/or GSH signalling (Nyathi & Baker 2006;Munné-Bosch et al. 2013).Unfortunately, the literature is veryscarce on the cellular mechanisms involved with pAPX inperoxisomes.

The peroxisomal APX isoenzymes are localized in mem-branes and turned in the direction of the cytosol (Yamaguchiet al. 1995; Bunkelmann & Trelease 1996; Mullen et al. 1999;del Río et al. 2002).Thus, it seems plausible to argue that theseenzymes could also act on the cytosolic H2O2 homeostasis andmediate the peroxisome-cytosol cross-talk. If this conditionoccurs under in vivo conditions, this proposed mechanismcould create new paradigms for peroxisomal H2O2 metabolismand its overall antioxidant role. It has been widely acceptedthat the H2O2 produced in peroxisomes may cross the mem-brane towards the cytosol and/or chloroplasts and mitochon-dria (del Río et al. 2002). This feature could alleviate theoxidative effects on the peroxisomes, especially under CATdeficiency and/or high photorespiration. Moreover, theseenzymes could also reduce H2O2 concentrations in concertwith cytosolic APX, which would avoid toxicity and maintainthe H2O2 at levels suitable for signalling.Thus, the deficiency ofpAPX could generate specific redox signals, which in turncould trigger signalling for singular redox responses. Thesemechanisms could be distinct and more effective than thosedisplayed by the NT rice plants, which might only involveclassical antioxidant pathways (Mittler et al. 2011).

Interestingly, the NT plants with CAT deficiency exhibitedhigher H2O2 concentrations, evaluated by the semi-quantitative DAB staining method in the leaf segments,which were directly correlated with total glutathione levelsand GSH content. Although DAB staining is a semi-quantitative method, in certain situations, it might be suitableto evaluate peroxide accumulation at the tissue level (Quevalet al. 2008). An accurate quantitative measurement of H2O2

in leaf tissues by chemical and enzymatic method can be

Figure 9. Changes in (a) electrolyte leakage, (b) lipidperoxidation and (c) maximum quantum yields of PSII in NT andAPX4 leaf segments of rice plants exposed to the control(200 μmol m−2 s−1), 10 mM AT or high light(1000 mol m−2 s−1) + 10 mM AT. Different lowercase lettersrepresent significant differences between the treatments withinlines (NT and APX4), and different capital letters representsignificant differences between the lines within treatments (control,AT and HL) at a confidence level of 0.05. The data are presentedas the means of four replicates (±sd) and were compared byTukey’s test. AT, 3-amino-1,2,4-triazole; APX, ascorbateperoxidise; High Light (HL); PSII, photosystem II; NT,non-transformed.



difficult, and the obtained results are frequently controversialand not easily interpreted physiologically (Queval et al.2011). To overcome this problem, authors have utilized theH2O2-mediated gene expression as an alternative method forevaluation of H2O2 accumulation in peroxisomes of plantsdeficient in CAT (Queval et al. 2009; Noctor et al. 2013).

Despite the total inhibition of CAT activity in rice leaves,the expression of OsCATA and OsCATB genes in responseto AT application were contradictory. The OsCATB tran-script amount was strongly up-regulated in silenced mutants,whereas in NT plants, the up-regulation was lower. Althoughthe protein encoded by the OsCATB gene is not consideredto be located in peroxisomes (Iwamoto et al. 2000; Mhamdiet al. 2012), our group previously demonstrated that theexpression of the OsCATB gene was strongly up-regulatedin rice leaves after exposure to high salinity(Menezes-Benavente et al. 2004). Contrary to what has beenreported in the literature, the OsCATC gene, which mostlikely encodes a class I protein for peroxisomes, was notup-regulated by salinity (Menezes-Benavente et al. 2004)even after several times of salt exposure. The response pre-sented by the OsCATB gene has been recurrent in our lab inthe presence of other abiotic stress. However, OsCATAexpression has been frequently down-regulated in rice leavesin response to abiotic stress.Although these questions are nota central focus of the current work, they require further studyto determine the metabolic roles of CAT isoforms inphotorespiration activity in rice.

GSH metabolism in peroxisomes under CAT deficiency isimportant. The GSSG levels in the NT plants were similar tothose indicated by pAPX mutants. These results suggest thatan intense de novo synthesis of GSH was more prominent inNT plants, favouring its redox state. Increases in the totalglutathione pool have been frequently found in other species,especially in the Arabidopsis mutant cat2, which lacks CAT2(Queval et al. 2009; Han et al. 2013; Gao et al. 2014). Thebiochemical and molecular mechanisms that control theincrease in GSH synthesis in plants in response to increasedH2O2 are still being debated (Queval et al. 2011; Han et al.2013; Noctor et al. 2013). GSH is the first line of antioxidantdefence under high peroxisomal H2O2 conditions because itis easily regenerated from GR activity and synthesized denovo by the rate-limiting γ-glutamylcysteine synthetase(γ-ECS) enzyme (Noctor et al. 2012). The GSSG pool is bio-chemically stable and contributes to the maintenance ofadequate GSH regeneration via GR activity (Mhamdi et al.2010; Noctor et al. 2012).

The mechanisms associated with increased photo-respiratory H2O2 and GSH oxidation in this study are unclearbecause the expression of GPX isoforms was up-regulatedmore intensely in the mutants, in which the H2O2 levels werelower than in the NT plants. In certain species, GPX mayutilize H2O2 as an electron acceptor and GSH as a reducingagent (Iqbal et al. 2006; Chang et al. 2009; Bonifacio et al.2011). The GSH oxidation in this reaction might partiallyexplain the lower concentrations of H2O2 and GSH in thepAPX mutants. However, evidences have suggested thatunder in vivo conditions, thioredoxins and organic

hydroperoxide are more important than H2O2 and GSH forGPX activity (Herbette et al. 2002; Iqbal et al. 2006). TheGST activity apparently was not important for consumingGSH in both genotypes, despite this enzyme/gene familybeing very large and functionally complex (Dixon et al.2011). However, other reactions dependent on GSH reduc-tion were not measured here, such as monodehydroascorbatereductase (MDHAR), peroxidase-type GST, GSH-dependent thioredoxin reductase and other reactionsbetween GSH and thiol-proteins (Dietz 2014). These reac-tions could explain the intense GSH oxidation, especially inthe NT plants, but further studies are required to elucidatethis point in rice.

The improved acclimation displayed by silenced plants wasassociated with lower levels of total glutathione, a GSHredox state and similar GSSG contents and, this profile wasassociated with lower photorespiratory H2O2 levels. Thus,these parameters alone are not suitable to explain the effec-tive antioxidant protection against oxidative stress inducedby CAT deficiency in rice, which has been reported forcertain species (Smith et al. 1985; Willekens et al. 1997; Hanet al. 2013). Contrary to what has been widely reported forother species, such as Arabidopsis and tobacco (Mhamdiet al. 2010; Foyer & Noctor 2013; Han et al. 2013), the ASCredox state was greatly altered by high photorespiratoryH2O2 in both genotypes, strongly suggesting that riceintensely utilize this antioxidant during oxidative stress gen-erated by high photorespiratory H2O2. In addition, GSH,H2O2 and ASC are considered to be powerful signalling mol-ecules involved with redox homeostasis. The differencesfound between the two genotypes for acclimatizing to highphotorespiratory H2O2 may be associated with the signallingof these molecules. Alternatively, these differences couldhave involved only changes in the H2O2 balance, which isrelated to the capacity of production and scavenging andinvolve classical and non-classical antioxidants pathways(rates of GO activity, photorespiration and photosynthesis).

Recent evidence obtained from GR KO Arabidopsismutants and other studies has demonstrated that a set ofGSH-dependent gene expression overlaps with the H2O2-dependent genes and that the signal transduction pathwayrelated to CAT deficiency is very complex and may involveGR expression and phytohormones, such as jasmonic acid,auxins and brassinosteroids (Mhamdi et al. 2010; Jiang et al.2012; Gao et al. 2014). H2O2 may be involved in the signallingto express peroxidases (Karpinski et al. 1997; Klein et al.2012), which could explain the differential expression ofcertain APX and GPX isoforms in the mutant and NT plants(Bonifacio et al. 2011). In addition, ASC could have acted tosignal several processes (Foyer & Noctor 2011), but its role asa signalling molecule in rice is unclear. The signalling mecha-nisms that involve ASC are not as well understood as thoseattributed to H2O2 and GSH (Foyer & Noctor 2011).

Broadly, the metabolic alterations that occurred in bothrice plants exposed to high photorespiratory H2O2 may haveutilized cross-talk mechanisms that might have involvedperoxisomes chloroplasts and mitochondria because thesecompartments are localized close to the cytosol and the



coordinated photorespiratory metabolism involves all ofthese organelles (del Río et al. 2002; Peterhansel & Maurino2011). For example, cytosolic H2O2 is a powerful signallingmechanism in the expression of several genes involved inphotosynthesis (Davletova et al. 2005). The pAPX mutantssuffered smaller effects related to CO2 assimilation, and thisprocess is the most important electron consumer from thephotochemistry phase, restricting H2O2 production in chloro-plasts (Carvalho et al. 2014). Moreover, GO is a fundamentalenzyme involved in H2O2 formation in peroxisomes duringphotorespiration (Zelitch et al. 2009).

The lower GO activity observed in the pAPX mutants wascorrelated with the glyoxylate levels and CAT activity, whichsuggests lower photorespiratory H2O2 production (Carvalhoet al. 2014; Lu et al. 2014). Under CAT inhibition, theglyoxylate content was remarkably decreased only in the NTplants, indicating that this change could be related to pAPXsilencing or downstream metabolic alterations. Peroxisomalglyoxylate might be consumed in the glyoxylate cycle inglyoxysomes,and during senescence and under certain abioticstresses, the conversion of leaf peroxisomes into glyoxysomesis possible (del Río et al. 2002). The pAPX mutants showedsignificantly less senescence than the NT plants, and thisresponse was associated with lower H2O2 levels (measured byDAB staining). It is widely known that H2O2 is involved insenescence (for a review, see Pintó-Marijuan & Munné-Bosch2014),and the deficiency of pAPX in rice plants may affect thisresponse by mechanisms that are not clear.

In conclusion, the peroxisomal APX knockdown riceleaves displayed an unexpected physiological acclimation tothe oxidative stress generated by a transient CAT deficiency.However, the mutants deficient in pAPX exhibited importantalterations in their oxidative and antioxidant metabolism,which could have generated a phenotype more capable ofcoping with high photorespiratory H2O2. In this context, thehigher CO2 photosynthetic assimilation and lower GO activ-ity in leaves could mitigate the H2O2 production inperoxisomes. Because the glyoxylate content was lower in themutants, this metabolite may be involved in photosynthesisand photorespiration modulation with H2O2 as a signallingmechanism. However, further studies are required to explainthe features exhibited by peroxisomal APX-silenced ricemutants in the presence of CAT deficiency.

ACKNOWLEDGMENTS

The authors are grateful to the Conselho Nacional deDesenvolvimento Científico e Tecnológico (CNPq) – Proc.486231-2012-7 for the financial support and to ProfessorRogério Margis for the pAPX gene construct. J.A.G.S. andM.M.P. are CNPq honoured researchers.

REFERENCES

Amako K., Chen G.X. & Asada K. (1994) Separate assays specific for ascor-bate peroxidase and guaiacol peroxidase and for the chloroplastic andcytosolic isozymes of ascorbate peroxidase in plants. Plant and Cell Physi-ology 35, 497–504.

Baker A.L. & Tolbert N.E. (1966) Glycolate oxidase (ferredoxin containingform). Methods in Enzymology 9, 339–340.

Beauchamp C. & Fridovich I. (1971) Superoxide dismutase: improved assayapplicable to acrylamide gels. Analytical Biochemistry 44, 2762–2787.

Blum A. & Ebercon A. (1981) Cell membrane stability as a measure ofdrought and heat tolerance in wheat. Crop Science 21, 43–47.

Bonifacio A., Martins M.O., Ribeiro C.W., Fontenele A.V., Carvalho F.E.L.,Margis-Pinheiro M. & Silveira J.A.G. (2011) Role of peroxidases in thecompensation of cytosolic ascorbate peroxidase knockdown in rice plantsunder abiotic stress. Plant, Cell & Environment 34, 1705–1722.

Bunkelmann J. & Trelease R. (1996) Ascorbate peroxidase. A prominentmembrane protein in oilseed glyoxysomes. Plant Physiology 110, 589–598.

Cakmak I. & Horst W.J. (1991) Effect of aluminum on lipid peroxidation,superoxide-dismutase, catalase and peroxidase activities in root-tips ofsoybean (Glycine max). Physiologia Plantarum 83, 463–468.

Carvalho F.E.L., Ribeiro C.W., Martins M.O., Bonifacio A., Staats C.C.,Andrade C.M.B., . . . Silveira J.A.G. (2014) Cytosolic APX knockdown riceplants sustain photosynthesis by regulation of protein expression related tophotochemistry, Calvin cycle and photorespiration. Physiologia Plantarum150, 632–645.

Caverzan A., Bonifacio A., Carvalho F.E.L., Andrade C.M.B., Passaia G.,Schünemann M., . . . Margis-Pinheiro M. (2014) The knockdown ofchloroplastic ascorbate peroxidases reveals its regulatory role in the photo-synthesis and protection under photo-oxidative stress in rice. Plant Science(Shannon, Ireland) 214, 74–87.

Chamnongpol S. & Willekens H. (1996) Transgenic tobacco with a reducedcatalase activity develops necrotic lesions and induces pathogenesis-relatedexpression under high light. The Plant Journal: For Cell and MolecularBiology 10, 491–503.

Chang C.C.C., Slesak I., Jordá L., Sotnikov A., Melzer M., Miszalski Z., . . .Karpinski S. (2009) Arabidopsis chloroplastic glutathione peroxidases play arole in cross talk between photooxidative stress and immune responses.Plant Physiology 150, 670–683.

Davletova S., Rizhsky L., Liang H., Shengqiang Z., Oliver D.J., Coutu J., . . .Mittler R. (2005) Cytosolic ascorbate peroxidase 1 is a central component ofthe reactive oxygen gene network of Arabidopsis. The Plant Cell 17, 268–281.

Dietz K.-J. (2014) Redox regulation of transcription factors in plantstress acclimation and development. Antioxidants and Redox Signaling.doi:10.1089/ars.2013.5672. [Epub ahead of print].

Dixon D., Steel P. & Edwards R. (2011) Roles for glutathione transferases inantioxidant recycling. Plant Signaling & Behavior 6, 1223–1227.

Flexas J., Ribas-Carbó M., Diaz-Espejo A., Galmés J. & Medrano H. (2007)Mesophyll conductance to CO2: current knowledge and future prospects.Plant, Cell & Environment 31, 602–621.

Foyer C.H. & Noctor G. (2009) Redox regulation in photosynthetic organisms.Antioxidants and Redox Signaling 11, 861–905.

Foyer C.H. & Noctor G. (2011) Ascorbate and glutathione: the heart of theredox hub. Plant Physiology 155, 2–18.

Foyer C.H. & Noctor G.D. (2013) Redox signaling in plants. Antioxidants andRedox Signaling 18, 2087–2090.

Gao X., Yuan H.-M., Hu Y.-Q., Li J. & Lu Y.-T. (2014) Mutation of Arabidopsiscatalase2 results in hyponastic leaves by changes of auxin levels. Plant, Cell& Environment 37, 175–188.

Giannopolitis C. & Ries S. (1977) Superoxide dismutases II. purification andquantitative relationship with water-soluble protein in seedlings. PlantPhysiology 59, 315–318.

Griffith O.W. (1980) Determination of glutathione and glutathione disulfideusing glutathione reductase and 2-vinylpyridine. Analytical Biochemistry106, 207–212.

Han Y., Chaouch S., Mhamdi A., Queval G., Zechmann B. & Noctor G. (2013)Functional analysis of Arabidopsis mutants points to novel roles forglutathione in coupling H2O2 to activation of salicylic acid accumulation andsignaling. Antioxidants and Redox Signaling 18, 2106–2121.

Havir E.A. & McHale N.A. (1987) Biochemical and developmental charac-terization of multiple forms of catalase in tobacco leaves. Plant Physiology84, 450–455.

Herbette S., Lenne C., Leblanc N., Julien J.-L., Drevet J.R. & Roeckel-DrevetP. (2002) Two GPX-like proteins from Lycopersicon esculentum andHelianthus annuus are antioxidant enzymes with phospholipidhydroperoxide glutathione peroxidase and thioredoxin peroxidase activ-ities. European Journal of Biochemistry 269, 2414–2420.



Hertwig B., Streb P. & Feierabend J. (1992) Light dependence of catalasesynthesis and degradation in leaves and the influence of interfering stressconditions. Plant Physiology 100, 1547–1553.

Hoagland D.R. & Arnon D.I. (1950) The Water-Culture Method for GrowingPlants Without Soil, p. 32. California Agricultural Experiment Station,Berkeley, CA, USA.

Iqbal A., Yabuta Y., Takeda T., Nakano Y. & Shigeoka S. (2006)Hydroperoxide reduction by thioredoxin-specific glutathione peroxidaseisoenzymes of Arabidopsis thaliana. The FEBS Journal 273, 5589–5597.

Iwamoto M., Higo H. & Higo K. (2000) Differential diurnal expression of ricecatalase genes: the 5′-flanking region of CatA is not sufficient for circadiancontrol. Plant Science (Shannon, Ireland) 151, 39–46.

Jiang Y.-P., Cheng F., Zhou Y.-H., Xia X.-J., Mao W.-H., Shi K., . . . Yu J.-Q.(2012) Cellular glutathione redox homeostasis plays an important role in thebrassinosteroid-induced increase in CO2 assimilation in Cucumis sativus.The New Phytologist 194, 932–943.

Karpinski S., Escobar C., Karpinska B., Creissen G. & Mullineaux P.M. (1997)Photosynthetic electron transport regulates the expression of cytosolicascorbate peroxidase genes in Arabidopsis during excess light stress. ThePlant Cell 9, 627–640.

Kavitha K., Venkataraman G. & Parida A. (2008) An oxidative and salinitystress induced peroxisomal ascorbate peroxidase from Avicennia marina:molecular and functional characterization. Plant Physiology and Biochem-istry 46, 794–804.

Kendall A., Keys A., Turner J., Lea P. & Miflin B. (1983) The isolation andcharacterisation of a catalase-deficient mutant of barley (Hordeum vulgareL.). Planta 159, 505–511.

Klein P., Seidel T., Stöcker B. & Dietz K.-J. (2012) The membrane-tetheredtranscription factor ANAC089 serves as redox-dependent suppressor ofstromal ascorbate peroxidase gene expression. Frontiers in Plant Science 3,1–13.

Koussevitzky S., Suzuki N., Huntington S., Armijo L., Sha W., Cortes D., . . .Mittler R. (2008) Ascorbate peroxidase 1 plays a key role in the response ofArabidopsis thaliana to stress combination. The Journal of Biological Chem-istry 283, 34197–34203.

Li Y.-J., Hai R.-L., Du X.-H., Jiang X.-N. & Lu H. (2009) Over-expression of apopulus peroxisomal ascorbate peroxidase (PpAPX) gene in tobacco plantsenhances stress tolerance. Plant Breeding 128, 404–410.

Lu Y., Li Y., Yang Q., Zhang Z., Chen Y., Zhang S. & Peng X.X. (2014)Suppression of glycolate oxidase causes glyoxylate accumulation that inhib-its photosynthesis through deactivating rubisco in rice. PhysiologiaPlantarum 150, 463–476.

Menezes-Benavente L., Teixeira F.K., Alvim Kamei C.L. & Margis-PinheiroM. (2004) Salt stress induces altered expression of genes encoding antioxi-dant enzymes in seedlings of a Brazilian indica rice (Oryza sativa L.). PlantScience (Shannon, Ireland) 166, 323–331.

Mhamdi A., Hager J., Chaouch S., Queval G., Han Y., Taconnat L., . . . NoctorG. (2010) Arabidopsis glutathione reductase1 plays a crucial role in leafresponses to intracellular hydrogen peroxide and in ensuring appropriategene expression through both salicylic acid and jasmonic acid signalingpathways. Plant Physiology 153, 1144–1160.

Mhamdi A., Noctor G. & Baker A. (2012) Plant catalases: peroxisomal redoxguardians. Archives of Biochemistry and Biophysics 525, 181–194.

Miki D., Itoh R. & Shimamoto K. (2005) RNA silencing of single and multiplemembers in a gene family of rice. Plant Physiology 138, 1903–1913.

Miller G., Suzuki N. & Rizhsky L. (2007) Double mutants deficient in cytosolicand thylakoid ascorbate peroxidase reveal a complex mode of interactionbetween reactive oxygen species, plant development. Plant Physiology 144,1777–1785.

Mittler R., Vanderauwera S., Suzuki N., Miller G., Tognetti V.B., VandepoeleK., . . . Van Breusegem F. (2011) ROS signaling: the new wave? Trends inPlant Science 16, 300–309.

Mullen R.T., Lisenbee C.S., Miernyk J.A. & Trelease R.N. (1999) Peroxisomalmembrane ascorbate peroxidase is sorted to a membranous network thatresembles a subdomain of the endoplasmic reticulum. The Plant Cell 11,2167–2185.

Munné-Bosch S., Queval G. & Foyer C.H. (2013) The impact of global changefactors on redox signaling underpinning stress tolerance. Plant Physiology161, 5–19.

Narendra S., Venkataramani S., Shen G., Wang J., Pasapula V., Lin Y., . . . ZhangH. (2006) The Arabidopsis ascorbate peroxidase 3 is a peroxisomalmembrane-bound antioxidant enzyme and is dispensable for Arabidopsisgrowth and development. Journal of Experimental Botany 57, 3033–3042.

Noctor G., Mhamdi A., Chaouch S., Han Y., Neukermans J., Marquez-GarciaB., . . . Foyer C.H. (2012) Glutathione in plants: an integrated overview.Plant, Cell & Environment 35, 454–484.

Noctor G., Mhamdi A., Queval G. & Foyer C.H. (2013) Regulating the redoxgatekeeper: vacuolar sequestration puts glutathione disulfide in its place.Plant Physiology 163, 665–671.

Nyathi Y. & Baker A. (2006) Plant peroxisomes as a source of signallingmolecules. Biochimica et Biophysica Acta 1763, 1478–1495.

Panchuk I.I., Volkov R.A. & Schöffl F. (2002) Heat stress- and heat shocktranscription factor-dependent expression and activity of ascorbateperoxidase in Arabidopsis 1. Plant Physiology l29, 838–858.

Peterhansel C. & Maurino V.G. (2011) Photorespiration redesigned. PlantPhysiology 155, 49–55.

Pintó-Marijuan M. & Munné-Bosch S. (2014) Photo-oxidative stress markersas a measure of abiotic stress-induced leaf senescence: advantages and limi-tations. Journal of Experimental Botany 65, 3845–57.

Queval G. & Noctor G. (2007) A plate reader method for the measurement ofNAD, NADP, glutathione, and ascorbate in tissue extracts: application toredox profiling during Arabidopsis rosette development. Analytical Bio-chemistry 363, 58–69.

Queval G., Issakidis-Bourguet E., Hoeberichts F.A., Vandorpe M., Gakière B.,Vanacker H., . . . Noctor G. (2007) Conditional oxidative stress responses inthe Arabidopsis photorespiratory mutant cat2 demonstrate that redox stateis a key modulator of daylength-dependent gene expression, and definephotoperiod as a crucial factor in the regulation of H2O2-induced cel. ThePlant Journal: For Cell and Molecular Biology 52, 640–657.

Queval G., Hager J., Gakière B. & Noctor G. (2008) Why are literature data forH2O2 contents so variable? A discussion of potential difficulties in thequantitative assay of leaf extracts. Journal of Experimental Botany 59, 135–146.

Queval G., Thominet D., Vanacker H., Miginiac-Maslow M., Gakière B. &Noctor G. (2009) H2O2-activated up-regulation of glutathione inArabidopsis involves induction of genes encoding enzymes involved incysteine synthesis in the chloroplast. Molecular Plant 2, 344–356.

Queval G., Jaillard D., Zechmann B. & Noctor G. (2011) Increased intracellularH2O2 availability preferentially drives glutathione accumulation in vacuolesand chloroplasts. Plant, Cell & Environment 34, 21–32.

del Río L.A., Corpas F.J., Sandalio L.M., Palma J.M., Gómez M. & Barroso J.B.(2002) Reactive oxygen species, antioxidant systems and nitric oxide inperoxisomes. Journal of Experimental Botany 53, 1255–1272.

Ribeiro C.W., Carvalho F.E.L., Rosa S.B., Alves-Ferreira M., Andrade C.M.B.,Ribeiro-Alves M., . . . Margis-Pinheiro M. (2012) Modulation of genesrelated to specific metabolic pathways in response to cytosolic ascorbateperoxidase knockdown in rice plants. Plant Biology 14, 944–955.

Rizhsky L., Hallak-Herr E., Van Breusegem F., Rachmilevitch S., Barr J.E.,Rodermel S., . . . Mittler R. (2002) Double antisense plants lacking ascorbateperoxidase and catalase are less sensitive to oxidative stress than singleantisense plants lacking ascorbate peroxidase or catalase. The Plant Journal:For Cell and Molecular Biology 32, 329–342.

Rosa S.B., Caverzan A., Teixeira F.K., Lazzarotto F., Silveira J.A.G.,Ferreira-Silva S.L., . . . Margis-Pinheiro M. (2010) Cytosolic APx knockdownindicates an ambiguous redox responses in rice. Phytochemistry 71, 548–558.

Schrader M. & Fahimi H.D. (2006) Peroxisomes and oxidative stress.Biochimica et Biophysica Acta 1763, 1755–1766.

Schreiber U., Bilger W. & Neubauer C. (1994) Chlorophyll fluorescence as anonintrusive indicator for rapid assessment of in vivo photosynthesis. InEcophysiology of Photosynthesis (eds E.D. Schulze & M.M. Caldwell), pp.49–70. Springer, Berlin.

Shigeoka S., Ishikawa T., Tamoi M., Miyagawa Y., Takeda T., Yabuta Y. &Yoshimura K. (2002) Regulation and function of ascorbate peroxidaseisoenzymes. Journal of Experimental Botany 53, 1305–1319.

Singh N., Mishra A. & Jha B. (2014) Over-expression of the peroxisomalascorbate peroxidase (SbpAPX) gene cloned from halophyte Salicorniabrachiata confers salt and drought stress tolerance in transgenic tobacco.Marine Biotechnology (New York, N.Y.) 16, 321–332.

Smith I., Kendall A., Keys A., Turner J. & Lea P. (1985) The regulation of thebiosynthesis of glutathione in leaves of barley (Hordeum vulgare L.). PlantScience (Shannon, Ireland) 41, 11–17.

Suzuki N., Miller G., Sejima H., Harper J. & Mittler R. (2013) Enhanced seedproduction under prolonged heat stress conditions in Arabidopsis thalianaplants deficient in cytosolic ascorbate peroxidase 2. Journal of ExperimentalBotany 64, 253–263.



Teixeira F.K., Menezes-Benavente L., Margis R. & Margis-Pinheiro M. (2004)Analysis of the molecular evolutionary history of the ascorbate peroxidasegene family: inferences from the rice genome. Journal of Molecular Evolu-tion 59, 761–770.

Teixeira F.K., Menezes-Benavente L., Galvão V.C., Margis R. &Margis-Pinheiro M. (2006) Rice ascorbate peroxidase gene family encodesfunctionally diverse isoforms localized in different subcellular compart-ments. Planta 224, 300–314.

Thordal-Christensen H., Zhang Z., Wei Y. & Collinge D.B. (1997) Subcellularlocalization of H2O2 accumulation in papillae and hypersensitive responseduring the barley-powdery mildew interaction. The Plant Journal: For Celland Molecular Biology 11, 1187–1194.

Upadhyaya N.M., Surin B., Ramm K., Gaudron J., Schunmann P.H.D., TaylorW., . . . Wang M.B. (2000) Agrobacterium-mediated transformation of Aus-tralian rice cultivars jarrah and amaroo using modified promoters andselectable markers. Australian Journal of Plant Physiology 27, 201–210.

Valentovicová K., Huttová J., Mistrík I. & Tamás L. (2009) Effect of abioticstresses on glutathione peroxidase and glutathione S-transferase activity inbarley root tips. Plant Physiology and Biochemistry 47, 1069–1074.

Van Aken O. & Whelan J. (2012) Comparison of transcriptional changes tochloroplast and mitochondrial perturbations reveals common and specificresponses in Arabidopsis. Frontiers in Plant Science 3, 1–18.

Vanderauwera S., Suzuki N., Miller G., van de Cotte B., Morsa S., Ravanat J.-L.,. . . Mittler R. (2011) Extranuclear protection of chromosomal DNA fromoxidative stress. Proceedings of the National Academy of Sciences of theUnited States of America 108, 1711–1716.

Wang J., Zhang H. & Allen R.D. (1999) Overexpression of an Arabidopsisperoxisomal ascorbate peroxidase gene in tobacco increases protectionagainst oxidative stress. Plant and Cell Physiology 40, 725–732.

Willekens H., Chamnongpol S., Davey M., Schraudner M., Langebartels C.,Van Montagu M., . . . Van Camp W. (1997) Catalase is a sink for H2O2 and isindispensable for stress defence in C3 plants. The EMBO Journal 16, 4806–4816.

Yamaguchi K., Mori H. & Nishimura M. (1995) A novel isoenzyme of ascor-bate peroxidase localized on glyoxysomal and leaf peroxisomal membranesin pumpkin. Plant and Cell Physiology 36, 1157–1162.

Yoshida K., Watanabe C., Kato Y., Sakamoto W. & Noguchi K. (2008) Influ-ence of chloroplastic photo-oxidative stress on mitochondrial alternative

oxidase capacity and respiratory properties: a case study with Arabidopsisyellow variegated 2. Plant and Cell Physiology 49, 592–603.

Zelitch I., Schultes N.P., Peterson R.B., Brown P. & Brutnell T.P. (2009) Highglycolate oxidase activity is required for survival of maize in normal air.Plant Physiology 149, 195–204.

Received 13 January 2014; received in revised form 2 July 2014;accepted for publication 3 July 2014

SUPPORTING INFORMATION

Additional Supporting Information may be found in theonline version of this article at the publisher’s web-site:

Figure S1. Habitus of representative plants from four repli-cates (each replicate represented by a pot containing twoplants). Non-transformed (NT) plants and three lines ofOsApx4-silenced plants (Lg, Lh and Lj) after 45 d of growthin a complete nutrient solution under normal growth condi-tions.Figure S2. Transcript levels of OsCat genes in NT andAPX4 leaves of rice plants exposed to control or 10 mMAT. Different lowercase letters represent significant differ-ences between the lines within treatments (control and AT),and different capital letters represent significant differencesbetween the treatments within lines (NT and APX4)at a confidence level of 0.05. The data are the meansof four replicates (±sd) and were compared using Tukey’stest.Figure S3. Time course of GO activity in the leaf segments ofNT rice plants exposed to the control or 10 mM HPMS over12 h. The data are the means of the four replicates (±sd).



Peroxisomal APX knockdown triggers antioxidant …Feb 06, 2014 · The physiological role of...

Documents

Transcript of Peroxisomal APX knockdown triggers antioxidant …Feb 06, 2014 · The physiological role of...